HIV-specific cytotoxic T-lymphocytes (CTL) play a critical role in containing HIV viremia in acute infection or in situation of spontaneous control, rendering them attractive candidates for vaccine strategies. However vaccines developed so far have failed to protect against HIV infection, this despite generating CTL responses. The fact that CTL responses against certain areas of the virus may be more effective at controlling HIV replication than CTL with other specificities suggests that vaccine should elicit selected CTL responses associated with protection. Effective presentation of the cognate epitopes by HIV-infected cells is a crucial condition for protective CTL responses. However, there is still surprisingly little understanding of intracellular mechanisms governing the presentation of HIV epitopes recognized by CD8 T cells. Most studies on HIV- specific CTL functions utilize cognate epitopes in the form of synthetic peptides, thus bypassing all intracellular steps of protein degradation leading to the presentation of epitopes. HIV infects several CD4-expressing subsets (CD4 T cells, monocyte/macrophages and dendritic cells) that will present HIV epitopes. Whether these subsets present similar epitopes with identical kinetics is unknown. Variations in epitope presentation between subsets may affect the antiviral efficacy of CTL. Conversely identifying areas of HIV that are efficiently processed into epitopes in all subsets is of highest importance for the identification of protective CTL responses and selection of immunogens. Building on novel epitope processing assays, we showed preferential processing of some HIV epitopes, a property that relies on motifs we used to alter the production of irrelevant epitopes. We also identified a novel factor involved in epitope processing efficiency, namely the highly variable intracellular stability of optimal HIV epitopes, also driven by specific motifs. Finally we show that CD4 T cells have lower processing activities than monocytes, which affects the kinetics and antigenicity of degradation products from HIV proteins. These data suggest that epitope production is controlled by rules that could be exploited to design customized immunogens. Specifically we propose to: 1) Determine whether CTL responses associated with spontaneous control of HIV viremia efficiently recognize and kill all HIV-infectable cell subsets. Taking advantage of a large cohort of controllers and progressors, we will assess the functionality of CD8 T cells stimulated by various HIV-infected subsets. 2) Identify peptides commonly produced in distinct antigen processing pathways of infectable subsets contributing to spontaneous controlled viremia. Using nanoparticles to target HIV proteins inside cell subsets, we will identify antigenic peptides produced by all subsets. 3) Design and test sequence signatures leading to the selective presentation of protective HIV epitopes. This proposal relies on a cross-disciplinary collaborative approach involving computational science, bioengineered tools and biochemical and immunological assays of epitope processing and CTL functions designed for primary cells. HIV-specific CD8 T cells are able to kill HIV-infected cells because infected cells display pieces of virus called epitopes. Deciphering the mechanisms of production and presentation of HIV epitopes stimulating efficient immune responses is key to rational vaccine design. This project seeks to identify immune responses and epitopes presented by HIV-infected cell subsets from persons with spontaneously undetectable viremia (controllers) and to define factors governing the efficiency of epitope processing that will enable selected presentation of protective epitopes in immunogens.